WSEAS Transactions on Systems and Control
Print ISSN: 1991-8763, E-ISSN: 2224-2856
Volume 21, 2026
Multi-Model System for Enhanced Road Safety: A Deep Learning Approach
Authors: , , , , ,
Search Articles
Abstract: In view of the increasing number of accidents and fatalities on the roads, the need for the application of automatic road safety surveillance has also increased. The paper proposes a multi-model system for improved road safety, wherein deep learning models, primarily based on Convolutional Neural Networks (CNN), are used for the enforcement of critical safety parameters. The parameters include helmet usage, seat belts, lane discipline, and face mask usage. The proposed research is based on the development of individual detection models. However, the models are made to run either individually or in parallel, thereby ensuring real-time checking for the enforcement of the aforementioned parameters in the presence of demanding vehicles and conditions. The proposed model, despite being based on real-time operation, has the advantage of ensuring ease of operation without the requirement for excessive computation, thus helping law enforcement agencies in pursuing violators for the aforementioned offenses.
Keywords:
Road safety, Real-time monitoring, Traffic control, Convolutional neural networks, Deep learning, Multi-model system
Pages: 112-121
DOI: 10.37394/23203.2026.21.12